SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition

We propose a self-training scheme, SURABHI, that trains deep-learning keypoint detection models on machine-annotated instances, together with the methodology to generate those instances. SURABHI aims to improve the keypoint detection accuracy not by altering the structure of a deep-learning-based ke...

Full description

Saved in:
Bibliographic Details
Main Authors: Manu Ramesh, Amy R. Reibman
Format: Article
Language:English
Published: MDPI AG 2024-11-01
Series:Sensors
Subjects:
Online Access:https://www.mdpi.com/1424-8220/24/23/7680
Tags: Add Tag
No Tags, Be the first to tag this record!
_version_ 1850260152859492352
author Manu Ramesh
Amy R. Reibman
author_facet Manu Ramesh
Amy R. Reibman
author_sort Manu Ramesh
collection DOAJ
description We propose a self-training scheme, SURABHI, that trains deep-learning keypoint detection models on machine-annotated instances, together with the methodology to generate those instances. SURABHI aims to improve the keypoint detection accuracy not by altering the structure of a deep-learning-based keypoint detector model but by generating highly effective training instances. The machine-annotated instances used in SURABHI are hard instances—instances that require a rectifier to correct the keypoints misplaced by the keypoint detection model. We engineer this scheme for the task of predicting keypoints of cattle from the top, in conjunction with our Eidetic Cattle Recognition System, which is dependent on accurate prediction of keypoints for predicting the correct cow ID. We show that the final cow ID prediction accuracy on previously unseen cows also improves significantly after applying SURABHI to a deep-learning detection model with high capacity, especially when available training data are minimal. SURABHI helps us achieve a top-6 cow recognition accuracy of 91.89% on a dataset of cow videos. Using SURABHI on this dataset also improves the number of cow instances with correct identification by 22% over the baseline result from fully supervised training.
format Article
id doaj-art-fe9d400080c84dd6b11dde803fcfe5a2
institution OA Journals
issn 1424-8220
language English
publishDate 2024-11-01
publisher MDPI AG
record_format Article
series Sensors
spelling doaj-art-fe9d400080c84dd6b11dde803fcfe5a22025-08-20T01:55:41ZengMDPI AGSensors1424-82202024-11-012423768010.3390/s24237680SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle RecognitionManu Ramesh0Amy R. Reibman1School of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907, USASchool of Electrical and Computer Engineering, Purdue University, West Lafayette, IN 47907, USAWe propose a self-training scheme, SURABHI, that trains deep-learning keypoint detection models on machine-annotated instances, together with the methodology to generate those instances. SURABHI aims to improve the keypoint detection accuracy not by altering the structure of a deep-learning-based keypoint detector model but by generating highly effective training instances. The machine-annotated instances used in SURABHI are hard instances—instances that require a rectifier to correct the keypoints misplaced by the keypoint detection model. We engineer this scheme for the task of predicting keypoints of cattle from the top, in conjunction with our Eidetic Cattle Recognition System, which is dependent on accurate prediction of keypoints for predicting the correct cow ID. We show that the final cow ID prediction accuracy on previously unseen cows also improves significantly after applying SURABHI to a deep-learning detection model with high capacity, especially when available training data are minimal. SURABHI helps us achieve a top-6 cow recognition accuracy of 91.89% on a dataset of cow videos. Using SURABHI on this dataset also improves the number of cow instances with correct identification by 22% over the baseline result from fully supervised training.https://www.mdpi.com/1424-8220/24/23/7680self-traininghard instanceskeypoint detectioncattle recognition
spellingShingle Manu Ramesh
Amy R. Reibman
SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition
Sensors
self-training
hard instances
keypoint detection
cattle recognition
title SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition
title_full SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition
title_fullStr SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition
title_full_unstemmed SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition
title_short SURABHI: Self-Training Using Rectified Annotations-Based Hard Instances for Eidetic Cattle Recognition
title_sort surabhi self training using rectified annotations based hard instances for eidetic cattle recognition
topic self-training
hard instances
keypoint detection
cattle recognition
url https://www.mdpi.com/1424-8220/24/23/7680
work_keys_str_mv AT manuramesh surabhiselftrainingusingrectifiedannotationsbasedhardinstancesforeideticcattlerecognition
AT amyrreibman surabhiselftrainingusingrectifiedannotationsbasedhardinstancesforeideticcattlerecognition